JetBrains launches Junie Local for offline AI coding on developer machines

Offline AI Coding with Junie Local

JetBrains has introduced Junie Local, a free version of its AI coding agent designed to operate entirely on a developer's local machine. This release addresses the demand for offline AI coding tools, allowing developers to maintain code on their own hardware, avoid per-request API costs, and work without an internet connection. Unlike many cloud-dependent AI coding tools, Junie Local provides a self-contained solution.

Simplified Local Model Setup

While other tools like Cline, Continue, and Aider support local runtimes such as Ollama or LM Studio, they often require developers to manually select models, configure quantization, and tune settings. Junie Local simplifies this process by pre-selecting and quantizing a model, and tuning its inference engine and agent harness for that specific combination. The setup is integrated into Junie itself; running the /local command downloads the necessary components, starts a local server, and automatically switches the agent.

Background on Junie

JetBrains, known for developer tools like IntelliJ IDEA, PyCharm, and WebStorm, initially launched Junie in January 2025 as an AI coding agent embedded in its IDEs. Junie is capable of planning tasks, modifying code, running tests, and working within a project's context. It has since expanded into a standalone command-line interface (CLI). Although Junie previously allowed connections to local runtimes, Junie Local streamlines the experience by providing a fully integrated, pre-configured offline solution.